Elective radical neck dissection in epidermoid cancer of the head and neck. A retrospective analysis of 853 cases of mouth, pharynx, and larynx cancer.

A retrospective analysis of 853 patients with cancer of the mouth, pharynx, and larynx operated on over a 30-year period was performed. Four hundred fifty-seven of them had a radical neck dissection (RND) at some point. Five hundred ninety patients had no clinically positive nodes (N-o) necks at the time of primary treatment; 99 of these had elective neck dissection, whereas 95 others had a delayed RND when nodes became clinically involved. Twenty-three percent of all N-o patients had microscopically involved nodes and less than half of these were among those patients selected for elective RND. Furthermore, 58% of those patients who had elective RND did not have positive nodes. Comparative analysis of elective RND, delayed therapeutic RND after clinical appearance of nodes, and composite operations for patients with N1-N3 disease indicates little difference in disease-free survival when the nodes in the elective RND were positive microscopically for tumors (56%, 49% and 47% respectively). It thus seems that elective RND offers no real advantage over a careful watchful waiting approach in most patients.